Background: Throughout the Syrian conflict, the Syrian government has intentionally attacked health facilities, violating International Humanitarian Law. Previous studies have qualitatively described health system disruptions following attacks on healthcare or established associations between armed conflict and health service utilisation, but there are no quantitative studies exploring the effects of health facility attacks. Our unprecedented study addresses this gap by quantifying the effects of health facility attacks on health service use during the Syrian conflict.

Methods: This retrospective observational study uses 18 537 reports capturing 2 826 627 consultations from 18 health facilities in northwest Syria and 69 attacks on these facilities. The novel study applies case time series design with a generalised non-linear model and stratification by facility type, attack mechanism and corroboration status.

Results: The study found significant, negative associations between health facility attacks and outpatient, trauma and facility births. On average, a health facility attack was associated with 51% and 38% reductions in outpatient, RR 0.49 (95% CI 0.43 to 0.57) and trauma consultations, RR 0.62 (95% CI 0.53 to 0.72), the day after an attack, with significant reductions continuing for 37 and 20 days, respectively. Health facility attacks were associated with an average 23% reduction in facility births, the second day after an attack, RR 0.77 (95% CI 0.66 to 0.89), with significant reductions continuing for 42 days.